Binding of DNA-Intercalating Agents to Oxidized ... - ACS Publications

Dec 4, 2015 - ABSTRACT: Quinone reductase 2 (NQO2) is an enzyme that might have intracellular signaling functions. NQO2 can exist in either an oxidize...
1 downloads 18 Views 21MB Size
Subscriber access provided by TUFTS UNIV

Article

The Binding of DNA Intercalating Agents to Oxidized and Reduced Quinone Reductase 2 Kevin Ka Ki Leung, and Brian Herbert Shilton Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.5b00884 • Publication Date (Web): 04 Dec 2015 Downloaded from http://pubs.acs.org on December 6, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

The Binding of DNA Intercalating Agents to Oxidized and Reduced Quinone Reductase 2

Kevin K.K. Leung† and Brian H. Shilton*

Department of Biochemistry, University of Western Ontario, 1151 Richmond St., London, Ontario, Canada N6A 5C1

*Corresponding Author Email: [email protected] Telephone: 519-661-4124 Fax: 519-661-3175 †

Present Address: Departments of Pharmaceutical Chemistry and Cellular & Molecular Pharmacology,

University of California at San Francisco.

1 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 38

Abstract Quinone reductase 2 (NQO2) is an enzyme that may have intracellular signalling functions. NQO2 can exist in either an oxidized or reduced state, and binding of compounds to one or both of these states inhibits enzymatic activity and could also affect intracellular signalling. A wide range of planar aromatic compounds bind NQO2, and we have identified three DNA intercalating agents – ethidium bromide, acridine orange (AO), and doxorubicin – as novel nanomolar inhibitors of NQO2. Ethidium and AO, which carry a positive charge in their aromatic ring systems, bound reduced NQO2 with 50fold higher affinity than oxidized NQO2, while doxorubicin bound only oxidized NQO2. Crystallographic analyses of oxidized NQO2 in complex with the inhibitors indicated that the inhibitors were situated deep in the active site. The aromatic faces were sandwiched between the isoalloxazine ring of FAD and the phenyl ring of F178, with their edges making direct contact with residues lining the active site. In reduced NQO2, ethidium and AO occupied a more peripheral position in the active site, allowing several water molecules to interact with the polar end of the negatively charged isoalloxazine ring. We also showed that AO inhibited NQO2 at a non-toxic concentration in cells while ethidium was less effective at inhibiting NQO2 in cells. Together, this study shows that reduced NQO2 has structural and electrostatic properties that yield a preference for binding of planar, aromatic, and positively charged molecules that can also function as DNA intercalating agents.

Keywords: Quinone Reductase 2, NQO2, DNA intercalation, ethidium bromide, acridine orange, doxorubicin, isothermal titration calorimetry, X-ray crystallography

2 ACS Paragon Plus Environment

Page 3 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Introduction Quinone Reductase 2 (NQO2) was historically classified as a detoxification enzyme responsible for the reduction of potentially cytotoxic quinones1. Similar to its sister enzyme Quinone Reductase 1 (NQO1), NQO2 catalyzes an obligate 2-electron transfer to generate the hydroquinone, thereby preventing production of reactive semi-quinones. Despite its clear relationship with NQO1 and related reductases, the cellular function of NQO2 is still an open question. In terms of potential endogenous substrates, NQO2 catalyzes reduction of Coenzyme-Q02; however, this quinone is primarily associated with the inner mitochondrial membrane as Coenzyme-Q10, and therefore it is most likely inaccessible to NQO2 in cells. It has been shown that NQO2 is able to efficiently catalyze the reduction of the ortho-quinone adrenochrome3 which is produced by the oxidation of adrenaline. Along similar lines, NQO2 was shown to bind and reduce estrogen ortho-quinones that are produced by oxidation of estrone or estradiol4. Estrogen and dopamine-derived ortho-quinones are also reduced by NQO15,6, and ortho-quinones undergo significant rates of non-enzymatic reduction by dihydronicotinamidecontaining co-enzymes4. Therefore, although ortho-quinones may represent endogenous cellular substrates for NQO2, it is not clear that reduction of ortho-quinones is the bona fide cellular function of NQO2. Whether NQO2 has a direct role in cellular detoxification is also open to debate. NQO1 clearly participates in cellular detoxification: it is induced by a variety of chemicals and is regulated along with many other detoxifying and cytoprotective genes as part of the Keap1/Nrf2/ARE pathway7. NQO2 is expressed in a number of different tissues8 but there is limited knowledge of how NQO2 expression is regulated. In an early report, NQO1 but not NQO2 mRNA was induced by 2,3,7,8-tetrachlorodibenzop-dioxin (TCDD)9. On the other hand, the NQO2 promotor harbours an antioxidant response element (ARE) and three xenobiotic response elements (XRE), and when coupled to a plasmid-borne CAT

3 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 38

reporter, β-napthoflavone (β-NF) and TCDD increased CAT activity 5- and 2-fold, respectively8. A later report showed a modest induction of NQO2 in Hep-G2 cells in response to β-NF, but this was in marked contrast to NQO1 which showed a robust response to β-NF and tert-butyl hydroquinone from the chromosomally encoded gene, and this response was recapitulated when the NQO1 promoter was used to drive a plasmid-borne CAT gene10. In summary, there is neither an obvious endogenous substrate for NQO2, nor a clear role in metabolism of xenobiotics and cellular detoxification. NQO2 exhibits an unusual preference for dihydronicotinamide riboside (NRH) and N-methyldihydronicotinamide as reducing coenzymes. For example, NQO1 efficiently catalyzes quinone reduction using NAD(P)H as a reducing coenzyme (kcat/KM = 440 min-1ŊµM-1), while NQO2 uses NAD(P)H inefficiently, with a catalytic efficiency 700-fold lower than that of NQO1 (kcat/KM = 0.62 min-1ŊµM-1)11. On this basis, it appears that NQO2 has evolved to avoid using the canonical reducing coenzymes NADH and/or NADPH and instead exhibits a preference for unusual, and possibly nonphysiological, reducing coenzymes, such as NRH11. Although nicotinamide riboside (NR) participates in NAD metabolism, it is currently unknown how the reduced form, NRH, is generated in cells. Indeed, the available evidence indicates that cells lack significant quantities of a suitable reducing co-enzyme for NQO212. Thus, NQO2 uses NAD(P)H inefficiently, lacks obvious physiologically relevant quinone substrates, and has a mode of regulation that is not clearly connected with cellular detoxification, which together suggest that the primary function of NQO2 is something other than the catalytic reduction of quinones. NQO2 has been implicated in cell signalling as well as the formation of reactive oxygen species13,14. Specifically, both NQO1 and NQO2 have been associated with regulation of the p53 tumour suppressor15,16 and there is evidence that NQO2 indirectly regulates the AKT/cyclin-D1/Rb and NFκB signalling pathways13,17. Numerous bioactive compounds interact with NQO2, raising the possibility

4 ACS Paragon Plus Environment

Page 5 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

that some of the cellular effects of the compounds may be attributed to NQO2. These compounds range from targeted kinase inhibitors (imatinib and nilotinib)18,19, CK2 inhibitors20, PKC inhibitors21, antimalarial compounds (primaquine, chloroquine, and quinacrine)22,23, quercetin11, resveratrol24,25, and melatonin26. All of these compounds can perturb cellular proliferation and/or apoptosis, although a specific role for NQO2 in the cellular effects of the compounds has not been determined. Nevertheless, the interaction of NQO2 with this diverse group of biologically active compounds, combined with observations that point towards a non-enzymatic role for NQO2, raise the possibility that NQO2 may function as a signalling molecule in the cell. Along these lines, reduction of NQO2 and chloroquine binding change the structure and/or dynamics of NQO227, suggesting that NQO2 could function as a flavin redox switch28 to signal changes in the cellular redox state along with the presence of biologically active and potentially cytotoxic molecules. The active site of NQO2 is composed of a large hydrophobic pocket with a planar aromatic surface provided by the isoalloxazine ring of the FAD co-factor29. As such, the planar aromatic portions of NQO2 inhibitors bind to the active site through Pi-Pi stacking interactions with the isoalloxazine ring and hydrophobic residues. Given the planar nature of NQO2 inhibitors, it is perhaps not surprising that a number of them, such as imatinib, quinacrine, and 9-aminoacridine, can also intercalate DNA30–32. In fact, several compounds known to intercalate DNA (9-aminoacridine and C1311) were identified by in silico screening for NQO2 inhibitors33,34. Given a potential role of NQO2 in cell signalling and the nature of compounds that bind NQO2, we investigated whether NQO2 interacts with wellcharacterized DNA intercalating agents. DNA intercalating agents are polycyclic aromatic molecules that can be mutagenic and cytotoxic to cells. In 1961, Leonard Lerman proposed and eventually showed that these compounds can unwind and extend DNA by inserting themselves between two nucleotide bases35. In highly proliferating cells, these

5 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 38

compounds are cytostatic by inhibiting DNA replication or transcription36,37. For cells that overcome replicative stress, the intercalating agents elongate DNA and cause DNA slippage leading to indel mutations during DNA replication38,39. When present in the environment, cells can deal with these harmful compounds by active excretion through efflux pumps such as P-glycoprotein (multidrug resistant-associated protein) and/or metabolism by cytochrome P45040; however, cells unable to clear DNA intercalating agents are at risk of cell growth retardation and genomic instability. Since DNA intercalating agents can be detrimental to cells, we hypothesize that NQO2 may have evolved to function, at least in part, as a sensor for the presence of potential DNA intercalating agents. In this study, we have discovered that three well-characterized DNA intercalating agents – ethidium bromide (EtBr), acridine orange (AO), and doxorubicin – are nanomolar inhibitors of NQO2. We further characterized these three novel inhibitors of NQO2 by means of enzyme kinetics, isothermal titration calorimetry, and crystallographic analysis. We found that the cationic ethidium and AO bound reduced NQO2 with a 50-fold higher affinity than oxidized NQO2; this can be explained by the increased negative electrostatic potential in the active site of reduced NQO2. In this regard, we have identified AO to be the highest affinity inhibitor for reduced NQO2, with a dissociation constant of 0.36 nM. Structurally, the inhibitors bind to the NQO2 active site in a manner that resembles their interaction with DNA. Lastly, we demonstrate that a non-toxic concentration of AO can inhibit NQO2 activity in cells.

6 ACS Paragon Plus Environment

Page 7 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Materials and Methods Reagents Recombinant NQO2 was expressed in E. coli and purified as previously described41. 1-(3sulfonatopropyl)-3-carbamoyl-1,4-dihydropyrimidine (SCDP) is an analogue of NRH obtained from Sigma-Aldrich. Nicotinamide riboside (NR) was purchased from High Performance Nutrition (Irving, CA, USA) and dihydronicotinamide riboside (NRH) was prepared according to previously described protocol with NR42. Ethidium bromide (EtBr) was obtained from Bioshop, acridine orange (AO) and doxorubicin were obtained from Caymen Chemical. Enzymatic Activity Measurements and Analysis Enzyme activity was measured by monitoring the consumption of SCDP at an absorbance peak of 360 nm using a Cary 100-Bio spectrophotometer (Varian). The reaction was initiated by addition of NQO2 to a final concentration of 154 pM in a stirred cuvette containing 150 μM SCDP and 5 μM menadione (Sigma) at 30°C with EtBr (1.9 nM to 2.6 µM), AO (7.8 nM to 1 µM) or doxorubicin (80 nM to 50 µM) in a buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl). IC50 values were calculated using a dose-response model and represented as relative inhibition. The constants for inhibition (KI values) of NQO2 by ethidium were determined using the same kinetics assay described for determination of IC50 values. For each inhibitor, kinetic assays were performed either with a constant SCDP concentration of 150 μM and a varying menadione concentration or with a constant menadione concentration of 5 μM and a varying SCDP concentration. NQO2 is subject to substrate inhibition by menadione, and inhibitors of NQO2 can be competitive against SCDP, menadione, or both. On this basis, we used an equation for steady-state ping-pong catalysis that includes a term for substrate inhibition by menadione (KI(Md)), as well as constants for inhibition against both the oxidized and reduced forms of the enzyme43: 7 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Vobs =

K

app M(SCDP)

Vmax [SCDP][Md] app [Md] + K M(Md) [SCDP] + [SCDP][Md]

app where K M(SCDP) = K M(SCDP) (1+

and

Page 8 of 38

app K M(Md) = K M(Md) (1+

[Md] [I] + ) K I(Md) K I(α )

(1)

[I] ) K I(β )

In Equation 1, [SCDP] and [Md] are the concentrations of SCDP and menadione respectively. KI(Md) is the inhibition constant that accounts for substrate inhibition. [I] is the concentration of inhibitor and KI(α) and KI(β) are the inhibition constants that modify the apparent Michaelis constants KM(SCDP) and KM(Md), respectively. That is, KI(α) is the constant that accounts for the change in KM(SCDP) in the presence of inhibitor; hence, it describes the component of inhibition that is competitive against SCDP. KI(β) is the constant that accounts for the change in KM(Md) in the presence of inhibitor, and it describes the component of inhibition that is competitive against menadione. Fixing the kinetic parameters of the uninhibited steady-state kinetics KM(SCDP) (142 µM), KM(Md) (8.07 µM), and KI(Md) (1.62 µM) with values previously determined43, Vmax, KI(α), and KI(β) were determined by globally fitting data to Equation 1 using nonlinear regression (Igor Pro version 6.34A) with the IC50 values as initial estimates. Isothermal Titration Calorimetry Isothermal titration calorimetry (ITC) was used to assess the direct binding of the inhibitors to NQO2. Titrations were performed using a MicroCal VP-ITC microcalorimeter and all data were analyzed using Origin 7.0 (MicroCal). Samples of concentrated NQO2 (700 µM in 50 mM Tris-HCl, 150 mM NaCl, pH 7.5) and concentrated inhibitors (20 to 60 mM in H2O or DMSO) were diluted into ITC buffer (50 mM sodium phosphate, 150 mM NaCl pH 7.55) to final concentrations of 8 µM and 80 to 260 µM respectively. In cases where inhibitor stocks were suspended in DMSO, an equal amount of DMSO was also added to the protein sample to match the buffer composition between the sample and

8 ACS Paragon Plus Environment

Page 9 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

titrant. All solutions were thoroughly degassed by stirring under vacuum. All titrations were performed at 25°C or 35°C with 3-10 µl injections spaced by 3 to 5 minute intervals. The integrated binding isotherms were corrected for the heat of dilution using data from titrations of the individual inhibitors into the respective matched buffers in the absence of NQO2. For the binding of inhibitors to oxidized NQO2, the 1.35 mL sample cell was filled with an 8 µM solution of NQO2 and the 250 µL syringe was filled with 80 µM EtBr, AO, or doxorubicin. All ITC titration data against oxidized NQO2 were fitted to a single-site binding model with MicroCal Origin7 software. Reduced NQO2 is subject to rapid oxidation27; however, the calorimeter provided a closed environment that limited diffusion of oxygen into the solutions. To titrate compounds against reduced NQO2, SCDP was added to degassed solutions of both NQO2 and the inhibitors to a final concentration of 500 µM. To measure the very high affinities of ethidium and AO towards reduced NQO2, competition (displacement) titrations were performed by adding 80 µM of EtBr or AO into 8 µM of NQO2 pre-mixed with 80 µM chloroquine (CQ) and 500 µM SCDP. The binding of CQ alone to reduced NQO2 was determined by titrating 160 µM CQ into 8 µM of NQO2 and 500 µM SCDP, and the binding isotherm was fitted to a single-site binding model. Using the resulting thermodynamic parameters of CQ binding to reduced NQO2, the binding isotherms of the competition titrations were fitted to a competitive binding model using MicroCal Origin7 software to determine the thermodynamic parameters for ethidium and AO. Crystal Structure Analysis Oxidized NQO2 (NQO2ox) was co-crystallized with ethidium, AO, or doxorubicin by hanging drop vapour diffusion against reservoirs containing 0.1 M HEPES pH 7.5, and 1.3 – 2.0 M (NH4)2SO4. To obtain crystals of reduced NQO2 in complex with ethidium or AO, reduction of NQO2ox-inhibitor crystals was performed as previously described27. Briefly, crystals of NQO2ox-ethidium or AO were

9 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 38

repeatedly soaked into 1 µL of reducing-soak solution with 0.1 M HEPES pH 7.5, 2.0 M (NH4)2SO4, 10 mM SCDP, and 1 mM of inhibitor, for 2 minute intervals until the crystals bleached. They were then transferred to a soak without SCDP before briefly passing through a cryoprotectant solution (2.0 M (NH4)2SO4, 0.1 Hepes pH 7.5, 20% glycerol) and plunged into liquid nitrogen. To prevent autooxidation of NQO2, the entire crystal mounting process from harvesting to cryo-cooling was performed under an anoxic atmosphere in a glove bag purged with N2. Crystallographic data were collected from a rotating anode source, processed using MOSFLM44, and merged using Scala45. The structures were solved by molecular replacement with PDB-ID 1QR2 as a starting model29; refinement was carried out using PHENIX46. All crystals had the same primitive orthorhombic space group and contained a dimer in the asymmetric unit; given the high resolution of the crystallographic data, NCS restraints were not used for any of the refinements. Topology files for ethidium and AO were generated using PHENIX.ELBOW46. The “butterfly bend” of the isoalloxazine ring was determined by simulated annealing with planar restraints removed, as previously described27. DNA binding pocket comparison The solvent accessible area in DNA or NQO2 binding pocket was calculated using POVME 2.047,48. Briefly, a contiguous area was generated from an arbitrary point in the binding site of the ethidium unwound DNA structure49 or reduced NQO2-ethidium structure for a 10 Å spherical radius. Dimensions of the binding pockets were then measured at arbitrary points along the edge of the binding pocket. Effect of AO on HCT116 Colon Carcinoma Cells To determine the cytotoxicity of AO, HCT116 cells were seeded at 1000 cells per well in 96-well plates and allowed to attach overnight. Cells were then treated with AO (0.014 to 230 µM) and grown for an additional 48 hrs before assaying using Sulforhodamine B (SRB; Sigma) according to

10 ACS Paragon Plus Environment

Page 11 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

established protocol50. Briefly, cells were fixed with trichloroacetic acid for 1 hr at 4°C and washed with water 3 times. Cells were then stained with 0.3% SRB in 1% acetic acid for 20 mins at room temperature and washed with 1% acetic acid 4 times. The dye was re-solubilized in 10 mM Tris base and absorbance was measured using at 560 nm (Victor multi-plate reader, Perkin Elmer). Cell growth in treated samples was normalized against controls that were not treated. Data were analyzed and plotted using Prism 6.0f (GraphPad Software Inc). To assay the ability of AO to inhibit NQO2, HCT116 cells were seeded at 1000 cells per well in 96well plates and allowed to attach overnight. Cells were then treated with 7.8 µM to 1000 µM CB1954 with 50 µM NRH, with 50 µM NRH and 78 nM AO, or without NRH. Cells were grown for an additional 48 hrs before being assayed for viability using MTT dye. Briefly, cells were treated with 1 mM of MTT dye and were incubated at 37˚C for 2-4 hrs. Media was completely removed and the dye was resolubilized in DMSO to be measured at 560 nm using a plate reader. Cell growth in treatment wells was normalized against cells with no treatment and data were analyzed using Prism 6.0f (GraphPad Software Inc).

11 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 38

Results DNA intercalating agents inhibit NQO2 Ashman and co-workers first characterized NQO2 in 1962 and showed that a small concentration (10 nM) of a number of anthracene derivatives inhibited its activity51. Since then, many other planar polycyclic aromatic compounds were identified as inhibitors of NQO2 including compounds that also intercalate DNA17,34,52. To characterize the interaction between NQO2 and DNA intercalating agents in greater detail, we investigated the ability of 5 different DNA intercalating molecules to inhibit NQO2 (Fig. 1). Three of the intercalators – ethidium bromide (EtBr), acridine orange (AO), and doxorubicin – exhibited IC50 values in the nanomolar range; the other two, mitoxantrone and methylene blue, showed relatively weak inhibition of NQO2 catalysis with IC50 values greater than 10 µM (Fig. 2A and Table 1). NQO2 catalyzes quinone reduction by a ping-pong mechanism: the dihydronicotinamide co-enzyme is first oxidized, transferring a hydride ion to the NQO2-bound FAD co-factor; the quinone substrate then binds to reduced NQO2, and electrons are transferred to produce the corresponding hydroquinone. Thus, NQO2 can exist in either an oxidized or reduced state, and inhibitors can be competitive towards the dihydronicotinamide co-enzyme or the quinone substrate. In a kinetic assay, this means that an inhibitor can raise the apparent KM for the dihydronicotinamide coenzyme or the quinone substrate. Using our previously established kinetic model (Equation 1), the terms KI(α) and KI(β) account for changes in the apparent KM of SCDP (an NRH analogue) and menadione (the quinone substrate) respectively43. Our kinetic characterization of EtBr inhibition towards NQO2 showed that competition with menadione was the dominant mode of NQO2 inhibition (KI(β) = 6.7 ± 1.2 nM versus KI(α) ≈ 500

± 2270 nM; Fig. 2B, Table 2). Note that the very large error associated with KI(α) (Table 2) indicates that binding of ethidium to oxidized NQO2 has no significant effect on the kinetics, and therefore the 12 ACS Paragon Plus Environment

Page 13 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

parameter is poorly determined. In other words, ethidium binds reduced NQO2 with much greater affinity compared to oxidized NQO2, and this interaction is responsible for inhibition of NQO2 catalysis. Binding of DNA intercalating agents to oxidized and reduced NQO2 To further characterize interactions between DNA intercalators and NQO2, the binding of ethidium, AO, and doxorubicin to both oxidized and reduced NQO2 was measured using Isothermal Titration Calorimetry (ITC). Of the three compounds, AO bound oxidized NQO2 with the highest affinity, almost 10-fold greater than both ethidium and doxorubicin (Table 3, Figure S1A-C). Preliminary titrations of AO and ethidium against reduced NQO2 indicated that their binding affinities were much higher compared to oxidized NQO2. To accurately characterize the very high affinity binding of both ethidium and AO to reduced NQO2, competition ITC was used53. In competition ITC, a compound with high binding affinity is titrated against the target protein in complex with a relatively low affinity ligand. In the case of NQO2, chloroquine (CQ) was used as the low-affinity ligand. A direct titration of reduced NQO2 with CQ yielded a KD value of 0.58 µM (Table 3, Figure S1G) similar to its inhibition constant (0.6 µM) determined previously using enzyme kinetics23. Consistent with the kinetics assays, ethidium bound most tightly to reduced NQO2, with a KD over 60 times lower compared to the oxidized form (Table 3, Figure S1D). The situation was similar for AO, which bound preferentially to reduced NQO2 with sub-nanomolar affinity (Table 3, Figure S1E). Doxorubicin bound to oxidized NQO2 with sub-micromolar affinity but there was no observable binding to reduced NQO2 (Table 3, Figure S1F). Crystal structures of NQO2 with DNA intercalating agents There are numerous structures of oxidized NQO2 in complex with a variety of inhibitors, but only two in which inhibitor binding to reduced NQO2 has been structurally characterized27,43. In both cases,

13 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 38

reduction of the FAD cofactor led to a striking change in the binding mode of the inhibitors. We have extended this comparative analysis by solving the high resolution crystal structures of oxidized and reduced NQO2 in complex with ethidium and AO, as well as doxorubicin in complex with oxidized NQO2 (Table S1). In all of the oxidized NQO2 inhibitor complexes, the inhibitors are deeply buried in the active site and sandwiched between the isoalloxazine ring of FAD and the phenyl ring of F178 (Fig. 3A, C, and E). As such, binding of the inhibitors to oxidized NQO2 includes common aromatic stacking interactions. In the structure of reduced NQO2 with ethidium (Fig. 3B) the overall orientation of ethidium has not changed, but it is positioned less deeply in the active site. The space vacated by the ethidium is filled with three water molecules that mediate hydrogen bonds between the amino group on the ethidium and N161 and G174 of NQO2. The situation is similar for AO: when bound to oxidized NQO2, AO is positioned deep in the active site, stacking over the oxygens of the isoalloxazine ring and making direct contact with N161 and G174 (Fig. 3C). However, AO has rotated and occupies a more peripheral location when bound to reduced NQO2 (Fig. 3D), and again the space next to N161 and G174 is filled with water molecules. The structures of ethidium and AO in complex with reduced NQO2 can be compared with the other two available structures of reduced NQO2 – one in complex with chloroquine (CQ)27 and other in complex with the protein kinase CK2 inhibitor DMAT43. Structures are also available for these compounds in complex with oxidized NQO2, allowing a general description of the difference in binding modes to reduced and oxidized NQO2. In all cases, the inhibitors bound to oxidized NQO2 such that they were positioned deep in the active site and made direct contact with N161 and G174. When bound to reduced NQO2, however, the inhibitors were positioned in a more peripheral location, and there were water molecules filling the space between N161, G174, and the inhibitors. It appears that the properties of the reduced isoalloxazine ring, which include a “butterfly bend” of approximately

14 ACS Paragon Plus Environment

Page 15 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

5° (Table S1 and Figure S2) and a negative charge that will be delocalized between N1 and O2, make the region above the isoalloxazine oxygens less suitable for the aromatic stacking interactions observed in the oxidized structures. Instead, the inhibitors move away from this region which is then occupied by polar, non-aromatic water molecules. The active site of NQO2 has a negative electrostatic potential that becomes much stronger when the isoalloxazine ring is reduced and carries a formal negative charge43. Both ethidium and AO are positively charged at neutral pH, which explains their preference towards the negatively charged FADH over neutral FAD. Doxorubicin also has a positive net charge, but the charge resides on nonaromatic portions of the molecule that are excluded from the active site (Fig. 3E). Therefore, a positive charge in the planar aromatic portions of NQO2 inhibitors appears to determine their preference for reduced NQO2. This is consistent with the two other NQO2 inhibitors, chloroquine and quinacrine, which both carry a positive charge in their aromatic portions and exhibit a marked preference for binding to reduced NQO223. NQO2 crystallizes with a dimer in the asymmetric unit, and inhibitors can sometimes adopt different orientations in the two crystallographically distinct binding sites. This was the case for both the NQO2ox-ethidium and NQO2red-ethidium structures where electron density for a second ethidium molecule in the B-chain active site became evident during refinement (Figure S3). The second ethidium interacts with the first by means of pi-pi planar stacking of their benzyl substituents, and the rest of the ethidium is loosely sandwiched between surface residues surrounding the B-chain active site and residues from a symmetry-related molecule. We believe that this mode of binding is induced primarily by crystal packing and likely irrelevant for NQO2 in solution. In contrast to ethidium, both AO and doxorubicin bound in exactly the same manner to both subunits of the NQO2 dimer. Comparison between intercalators binding to DNA and NQO2

15 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 38

The high affinity binding of DNA intercalating agents to NQO2 prompted us to compare the binding pocket in unwound (intercalated) DNA with the active site of NQO2. Molecules that intercalate DNA create a cavity with pi-pi stacking interactions mediated by the bases and the potential for hydrogen bonds to the DNA backbone, as shown in the structure of ethidium bound to DNA49 (Figure 4A). The binding pocket in the DNA can accommodate molecules up to 13.5 Å in length at the major groove side and 7.9 Å in length at the minor groove side. Since the DNA is unwound by ethidium, the adjacent base pairs are not parallel, with the space between them varying between 2.5 Å to 3.7 Å. In NQO2, the isoalloxazine ring in the bottom of the pocket provides pi-pi stacking interactions with one face of the inhibitor, while hydrophobic and aromatic residues at the top of the pocket mediate additional interactions with the other planar face of the inhibitor (Figure 4B). Similar to the binding pocket in DNA, the NQO2 active site can accommodate molecules up to 13.9 Å in length and 2.7 Å in height. When the reduced NQO2-ethidium structure was superimposed to the known DNA-ethidium structure by aligning the coordinates of ethidium, the many residues of the NQO2 binding pocket also superimposed with the unwound DNA as shown by the overlap of the molecular surfaces of the two molecules (Figure 4C). In the reduced NQO2-AO to DNA-AO54 comparison, the positively charged nitrogenous base of AO makes two hydrogen bonds to the hydroxyl group of the two adjacent guanosine nucleotides via water molecules, in addition to pi-pi stacking interactions (Figure 4D). Similarly in NQO2, the nitrogenous base of AO is anchored to E193 by hydrogen bonds via a water molecule. Therefore, the active site of NQO2 bears a surprising resemblance to the intercalator binding pocket in unwound DNA. Even though the active site of NQO2 bears some resemblance to the binding pocked in unwound (intercalated) DNA, the affinities of DNA intercalators for NQO2 do not correlate well with their affinities for DNA. For example, of the agents studied in this manuscript, doxorubicin binds DNA most

16 ACS Paragon Plus Environment

Page 17 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

tightly (KD = 384 nM)55, while binding of ethidium and AO is much weaker (KD values of 15 and 36 µM, respectively)56. In the case of NQO2, ethidium and AO bind reduced NQO2 with very high affinity, with dissociation constants that are 4 to 5 orders of magnitude lower (i.e. 3.47 and 0.57 nM; Table 1) than those for their binding do DNA; doxorubicin, on the other hand, displays a relatively modest affinity for NQO2, roughly the same as its affinity for DNA. Furthermore, the other DNA intercalators, methylene blue and mitoxantrone, which did not display observable inhibition of NQO2, bind DNA with similar affinity as ethidium and AO56,57 In summary, even though there are similarities between the NQO2 active site and the space in DNA occupied by intercalators, NQO2 is not a completely faithful mimic of DNA. Inhibition of NQO2 by AO in cells Beyond an in vitro characterization of these novel inhibitors of NQO2, we were also interested in whether AO, to our knowledge the highest affinity inhibitor of reduced NQO2 characterized to date, could inhibit NQO2 in cells. First, we established the cytotoxicity of AO in HCT116 cells. HCT116 cells are colorectal carcinoma cells with a high level of NQO2 expression and activity52; in addition, these cells do not express P-glycoprotein58, preventing active efflux of AO and making them a good model to assay in-cell inhibition of NQO2 by AO. We found that AO inhibited HCT116 cell proliferation with an IC50 value of 3.9 ± 0.5 µM and on this basis concentrations of AO below approximately 500 nM do not affect cell proliferation (Figure 5A). To determine whether NQO2 is inhibited by AO in cells, we used the cancer prodrug CB1954 that is specifically activated by NQO2 in the presence of NRH12. CB1954 is a DNA alkylating agent that can be reduced to a much more toxic DNA cross-linking agent by NQO2 in the presence of NRH. Without NRH, the cytotoxicity of CB1954 was 348 ± 41 µM, and in the presence of 50 µM NRH, the cytotoxicity of CB1954 increased 20-fold to 17.8 ± 1.4 µM (figure 5B). Addition of a non-toxic dose of

17 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 38

AO (78nM) was able to partially reverse the NRH-dependent activation of CB1954 and lower its cytotoxicity in the presence of NRH to 114 ± 15 µM (Figure 5C). We also tested whether EtBr was able to inhibit NQO2 with a non-toxic concentration and found that EtBr had an IC50 value of 9.1 ± 0.9 µM. The highest concentration of EtBr tested (312nM) was only able to slightly reverse the activation of CB1954 by NRH (data not shown). Therefore, a non-toxic concentration of AO can effectively inhibit NQO2 in cells while EtBr is less effective at inhibiting NQO2 in cells.

18 ACS Paragon Plus Environment

Page 19 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Discussion We have identified and characterized the binding of three well-known DNA intercalating agents to NQO2. Doxorubicin (Adriamycin) is a widely used cancer chemotherapeutic and binds to NQO2 with roughly the same affinity as it binds to DNA. Doxorubicin is an anthraquinone and could therefore be a substrate for NQO2. However, the major metabolites of doxorubicin and closely-related duanorubicin are not reduced at the quinone functional group59,60. Furthermore, when NQO2 was incubated with 100 µM doxorubicin and 150 µM SCDP, there was no detectable oxidation of SCDP, indicating that NQO2 cannot catalyze reduction of doxorubicin under these in vitro conditions. Thus, NQO2 is inhibited by doxorubicin and is not involved in its metabolism. Positively charged polycyclic compounds are thought to be good DNA intercalators because they are recruited to the negatively charged phosphate backbone prior to intercalating DNA. Here we have shown that when the positive charge resides in the aromatic ring system of compounds such as ethidium and AO, they exhibit a marked preference for binding to reduced NQO2, with affinities towards reduced NQO2 that are over 50-fold greater compared to those for oxidized NQO2. In contrast, a positive charge outside the planar aromatic portions of the inhibitor, as in the case of doxorubicin, does little to enhance binding affinity for reduced NQO2. The results with ethidium and AO are in line with those of two other NQO2 inhibitors, chloroquine and quinacrine, that bind preferentially to reduced NQO2 and carry a positive charge in their aromatic ring systems23. In summary, the available data indicate that a positive charge in the aromatic ring system of an NQO2 inhibitor will enhance preferential binding to the reduced form of the protein. This raises the question of how NQO2 may be reduced in vivo. NQO2 uses NAD(P)H with a catalytic efficiency over three orders of magnitude lower than for NRH, and on this basis it appears that NQO2 has evolved to avoid using NAD(P)H and/or to favour the unconventional co-enzyme, NRH.

19 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 38

Paradoxically, the available evidence indicates that NRH is present, if at all, in only relatively small amounts: in human cells the cancer prodrug and DNA alkylating agent CB1954 is metabolized by NQO2 to a more toxic DNA cross-linking agent, but only in the presence of exogenous NRH42. On the basis of its inefficient use of NAD(P)H and the observation that NQO2 does not efficiently metabolize CB1954 without exogenous NRH, it can be argued that the primary function of NQO2 is not the catalytic reduction of quinones or other electrophiles. Instead, NQO2 may act as a redox signalling system, which would not necessarily require a continuously refreshed pool of reducing coenzyme, but instead a metabolic condition in which the reduction of NQO2 is enhanced sufficiently to allow it to propagate a signal. This could occur if changes in cellular metabolism resulted in the production of NRH. For example, Williams-Ashman and co-workers, who originally discovered NQO2, suggested that cellular NRH could arise in cells through breakdown of NAD(P)H61. Changes in the levels of nicotinamide coenzymes may coincide with cell stresses such as DNA damage that increases NAD consumption by poly-ADP-ribose polymerases62. Another scenario for redox-dependent signalling is that NQO2 may be reduced inefficiently by cellular NAD(P)H such that it is maintained in a primarily oxidized state, but the lifetime of the reduced state could be altered by interactions that hinder oxidation. For instance, molecules that bind with very high affinity to reduced NQO2 (such as AO and ethidium) are expected to stabilize the reduced form through thermodynamic and kinetic effects. NQO2 has been implicated in neurodegenerative disease and cancer tumorigenesis63,64, although the cellular functions of NQO2 remain a matter of debate. NQO2 knockout mice appear to have lower levels of p53, a weaker induction of p53 by cell stressors, and decreased apoptosis64,65. In addition, NQO2 has been repeatedly identified as a target for a variety of bioactive compounds, such as resveratrol24, melatonin26, ammosamide B66, antimalarials23,27, and kinase-targeted therapeutics20,43. Despite these provocative observations, the knockout or inhibition of NQO2 does not bring about acute

20 ACS Paragon Plus Environment

Page 21 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

cellular effects. For example, Karen Nolan and coworkers found that imidazoacridine-6-one compounds are potent NQO2 inhibitors but also intercalate DNA at similar concentrations34. An Noxide modification of the parent compound reduced its ability to intercalate DNA and lowered its cytotoxicity 20-fold while retaining its affinity towards NQO267. In other work, inhibition or siRNAmediated knockdown of NQO2 was linked to regulation of NF-κB, and a variety of NQO2 inhibitors were shown to be active in cells, but their toxicity was not correlated with their inhibition of NQO217. Similarly, we have shown concentrations of AO that are sufficient to inhibit NQO2 have no obvious effect on cell proliferation. It appears that the function of NQO2 becomes manifest only when cells are challenged by some other insult. Although this idea is supported by the unusual co-enzyme dependence of NQO2, it greatly complicates our ability to understand the exact cellular function of NQO2.

21 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 38

Tables Table 1. Inhibition of NQO2 by DNA Intercalating Agents Intercalating Agent

IC50 (nM)

Ethidium Bromide

11.8 ± 0.4

Acridine Orange

24.6 ± 4.1

Doxorubicin

254 ± 41.7

Mitoxantrone

> 10 000

Methylene Blue

> 10 000

Table 2. Steady-state kinetic parameters for inhibition of NQO2 by EtBr. Vmax1

1184 ± 30 s-1

KI(α) 1,3

555 ± 2270 nM

KI(β) 1,3

6.7 ± 1.24 nM

Km (SCDP)2

142 µM

Km (menadione)2

1.62 µM

Ki (menadione)2

8.07 µM

1

Calculated by globally fitting data to Equation 1 as described in Methods.

2

Steady state kinetic parameters for the uninhibited reaction were determined previously43

3

KI(α) and KI(β) are the constants for inhibition against SCDP and menadione, respectively.



22 ACS Paragon Plus Environment

Page 23 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Table 3. Binding of NQO2 Inhibitors to Oxidized and Reduced NQO2 NQO2

Inhibitor

KD (nM)

ΔG (kcal/mol)

ΔH (kcal/mol)

-TΔS (kcal/mol)

215 ± 28

-9.4

-7.8 ± 0.1

-1.6

AO

29.4 ± 14.5

-10.3

-7.6 ± 0.2

-2.6

Doxorubicin

274 ± 42

-8.9

-8.1 ± 0.2

-0.8

EtBr1

3.47 ± 0.30

-11.5

-6.7 ± 0.1

-4.8

AO1

0.36 ± 0.09

-12.9

-6.5 ± 0.1

-6.3

Doxorubicin2

NB

-

-

-

CQ

578 ± 97

-8.5

-17.7 ± 0.5

9.2

Oxidized EtBr

Reduced

1

Thermodynamic parameters were determined by competition titrations against NQO2 in complex with CQ as described in Methods. 2

There was no observable binding of doxorubicin to reduced NQO2 (Figure S1 F).

23 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 38

Figure Legends Figure 1. Chemical structures of the DNA intercalating agents used in this study. Figure 2. Ethidium bromide, acridine orange, and doxorubicin are inhibitors of NQO2. (A) Determination of IC50 values for EtBr (˜), AO (¢), and doxorubicin (p) against NQO2-catalyzed reduction of menadione. The IC50 values are listed in Table 1. (B) Determination of kinetic inhibition constants for EtBr. The inhibition of NQO2 by EtBr was characterized using a constant menadione concentration (5 μM) and varying SCDP (top), and a constant SCDP concentration (150 µM) with varying menadione (bottom). The enzymatic activity was measured in the absence of EtBr (˜) and at EtBr concentrations of 2 nM (¢), 10 nM (p), 50 nM (q), and 250 nM (!). The data were globally fitted to Equation 1 as described in Materials and Methods. The kinetic constants are summarized in Table 2. Figure 3. Binding of ethidium, acridine orange, and doxorubicin to NQO2. The inhibitors are sandwiched between the FAD co-factor (below the plane of inhibitor) and F178 (above the plane of the inhibitor) in oxidized NQO2 (A, C, E), and the inhibitors are excluded from this region by water molecules in reduced NQO2 (B, D). (A) Ethidium is deeply buried in the active site of oxidized NQO2 making hydrogen bonds with N161 directly and to D117, T71, and G68 via two water molecules. (B) Ethidium is less buried in reduced NQO2 and makes hydrogen bonds with N161 and G174 via three water molecules on the left side of the binding site, and with G68 and T71 directly on the right side of the binding site. (C) Acridine orange binds oxidized NQO2 by making hydrogen bonds to N161 via two water molecules. (D) In the reduced structure of NQO2 in complex with AO, AO makes one hydrogen bond to E193 via a water molecule. (E) The planar moiety of doxorubicin is inserted into the left side of the binding cleft where it is anchored by a hydrogen bond to N161. All electron density

24 ACS Paragon Plus Environment

Page 25 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

maps surrounding the inhibitors are Fc-Fo omit maps generated after three rounds of simulated annealing and contoured at 3 σ. Figure 4. Comparison of the inhibitor binding sites in reduced NQO2 and DNA. (A-B) Solvent accessible area (green blob) of the unwound DNA in the DNA-ethidium structure (A; Cambridge Structure Database ID ETHUAD10) and the NQO2 active site (B) viewed from the front (left panel) showing the hydrophobic and polar regions of the binding sites. Views from the top (middle), and from the side (right panel) show the estimated dimensions in the binding sites. (C) Molecular surface of reduced NQO2 bound to ethidium (left), DNA intercalated with ethidium (middle), and superimposition of ethidium (right) to compare the molecular surfaces of the NQO2 and DNA ethidium binding sites. (D) Molecular surface of reduced NQO2 bound to AO (left) showing that the nitrogenous base of AO makes a hydrogen bond to E193 via a water molecule; DNA intercalated with AO (middle, CSD ID ACCYGA10) showing that the nitrogenous base of AO makes a hydrogen bond with the hydroxyl group of two adjacent guanosine bases via water molecules; and superimposition of AO (right) showing the overlap of molecular surfaces between NQO2 and DNA.

Figure 5. NQO2 is inhibited by AO in HCT116 cells. (A) IC50 of AO in cells. HCT116 cells were treated with AO at the indicated concentration for 48 hours before being assayed by SRB. Cell growth of the treated cells was normalized to the cell growth of untreated cells and the data were fitted to a dose-response curve to calculate an IC50 for AO of 3.9 ± 0.5 µM. (B) Induction of CB1954 cytotoxicity by NRH. Cells were treated with the indicated concentration of CB1954 with (¢) and without (˜) 50 µM NRH for 48 hours before being tested for viability using an MTT assay. (C) Cellular inhibition of NQO2 by AO. Induction of CB1954 cytotoxicity with 50 µM NRH was partially reversed by co-

25 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 38

treatment with 78 nM AO (p). Cytotoxicity of CB1954 with (¢) and without (˜) 50 µM NRH are shown for reference in grey.

26 ACS Paragon Plus Environment

Page 27 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ABBREVIATIONS AO, acridine orange; CQ, chloroquine; CSD, Cambridge Structural Database; DMAT, 2dimethylamino-4,5,6,7-tetrabromo-1H-benzimidazole; EtBr, ethidium bromide; ITC, isothermal titration calorimetry; Md, menadione; NQO2, quinone reductase 2; NR, nicotinamide riboside; NRH dihydronicotinamide riboside; SCDP, 1-(3-sulfonatopropyl)-3-carbamoyl-1,4-dihydropyrimidine;

SRB, sulforhodamine-B.

Supporting Information The following supporting information is provided: Table S1. Crystallographic Statistics Figure S1. Binding of DNA intercalating agents to NQO2 characterized by isothermal titration calorimetry. Figure S2. Bending of the isoalloxazine ring system in crystal structures from the current study. Figure S3. Second binding site of ethidium to oxidized NQO2 (A) and reduced NQO2 (B). The supporting information is available in the PDF file “Leung-Shilton_Supporting-Information.pdf”.

Funding Sources This research was funded by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada. Kevin Leung was supported by Ontario Graduate Scholarships and a Canadian Institutes of Health Research Frederick Banting and Charles Best Canada Graduate Scholarship Doctoral Award.

27 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 38

References (1) Zhao, Q., Yang, X. L., Holtzclaw, W. D., and Talalay, P. (1997) Unexpected genetic and structural relationships of a long-forgotten flavoenzyme to NAD(P)H:quinone reductase (DT-diaphorase). Proc. Natl. Acad. Sci. U. S. A. 94, 1669–1674. (2) Boutin, J., Chatelain-Egger, F., Vella, F., Delagrange, P., and Ferry, G. (2005) Quinone reductase 2 substrate specificity and inhibition pharmacology. Chem. Biol. Interact. 151, 213–228. (3) Fu, Y., Buryanovskyy, L., and Zhang, Z. (2008) Quinone reductase 2 is a catechol quinone reductase. J. Biol. Chem. 283, 23829–23835. (4) Gaikwad, N. W., Yang, L., Rogan, E. G., and Cavalieri, E. L. (2009) Evidence for NQO2-mediated reduction of the carcinogenic estrogen ortho-quinones. Free Radic. Biol. Med. 46, 253–262. (5) Gaikwad, N. W., Rogan, E. G., and Cavalieri, E. L. (2007) Evidence from ESI-MS for NQO1catalyzed reduction of estrogen ortho-quinones. Free Radic. Biol. Med. 43, 1289–1298. (6) Segura-Aguilar, J., and Lind, C. (1989) On the mechanism of the Mn3(+)-induced neurotoxicity of dopamine:prevention of quinone-derived oxygen toxicity by DT diaphorase and superoxide dismutase. Chem. Biol. Interact. 72, 309–324. (7) Dinkova-Kostova, A. T., and Talalay, P. (2010) NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1), a multifunctional antioxidant enzyme and exceptionally versatile cytoprotector. Arch. Biochem. Biophys. 501, 116–123. (8) Jaiswal, A. K. (1994) Human NAD(P)H:quinone oxidoreductase2. Gene structure, activity, and tissue-specific expression. J. Biol. Chem. 269, 14502–14508. (9) Jaiswal, A. K., Burnett, P., Adesnik, M., and McBride, O. W. (1990) Nucleotide and deduced amino acid sequence of a human cDNA (NQO2) corresponding to a second member of the NAD(P)H:quinone oxidoreductase gene family. Extensive polymorphism at the NQO2 gene locus on chromosome 6. Biochemistry 29, 1899–1906. (10) Radjendirane, V., and Jaiswal, A. K. (1999) Coordinated induction of the c-jun gene with genes encoding quinone oxidoreductases in response to xenobiotics and antioxidants. Biochem. Pharmacol. 58, 597–603. (11) Wu, K., Knox, R., Sun, X. Z., Joseph, P., Jaiswal, a K., Zhang, D., Deng, P. S., and Chen, S. (1997) Catalytic properties of NAD(P)H:quinone oxidoreductase-2 (NQO2), a dihydronicotinamide riboside dependent oxidoreductase. Arch. Biochem. Biophys. 347, 221–228. (12) Knox, R. J., Jenkins, T. C., Hobbs, S. M., Chen, S., Melton, R. G., and Burke, P. J. (2000) Bioactivation of 5-(Aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) by Human NAD(P)H Quinone Oxidoreductase 2: A Novel Co-substrate-mediated Antitumor Prodrug Therapy. Cancer Res. 1954, 4179–4186. (13) Hsieh, T., Yang, C.-J., Lin, C.-Y., Lee, Y.-S., and Wu, J. M. (2012) Control of stability of cyclin D1 by quinone reductase 2 in CWR22Rv1 prostate cancer cells. Carcinogenesis 33, 670–677. (14) Reybier, K., Perio, P., Ferry, G., Bouajila, J., Delagrange, P., Boutin, J. A., and Nepveu, F. (2011) Insights into the redox cycle of human quinone reductase 2. Free Radic. Res. 45, 1184–1195. 28 ACS Paragon Plus Environment

Page 29 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

(15) Khutornenko, A. A., Roudko, V. V, Chernyak, B. V, Vartapetian, A. B., Chumakov, P. M., and Evstafieva, A. G. (2010) Pyrimidine biosynthesis links mitochondrial respiration to the p53 pathway. Proc. Natl. Acad. Sci. U. S. A. 107, 12828–12833. (16) Asher, G., Lotem, J., Cohen, B., Sachs, L., and Shaul, Y. (2001) Regulation of p53 stability and p53-dependent apoptosis by NADH quinone oxidoreductase 1. Proc. Natl. Acad. Sci. U. S. A. 98, 1188–1193. (17) Nolan, K. A., Dunstan, M. S., Caraher, M. C., Scott, K. A., Leys, D., and Stratford, I. J. (2012) In silico screening reveals structurally diverse, nanomolar inhibitors of NQO2 that are functionally active in cells and can modulate NF-κB signaling. Mol. Cancer Ther. 11, 194–203. (18) Rix, U., Hantschel, O., Dürnberger, G., Remsing Rix, L. L., Planyavsky, M., Fernbach, N. V, Kaupe, I., Bennett, K. L., Valent, P., Colinge, J., Köcher, T., and Superti-Furga, G. (2007) Chemical proteomic profiles of the BCR-ABL inhibitors imatinib, nilotinib, and dasatinib reveal novel kinase and nonkinase targets. Blood 110, 4055–4063. (19) Bantscheff, M., Eberhard, D., Abraham, Y., Bastuck, S., Boesche, M., Hobson, S., Mathieson, T., Perrin, J., Raida, M., Rau, C., Reader, V., Sweetman, G., Bauer, A., Bouwmeester, T., Hopf, C., Kruse, U., Neubauer, G., Ramsden, N., Rick, J., Kuster, B., and Drewes, G. (2007) Quantitative chemical proteomics reveals mechanisms of action of clinical ABL kinase inhibitors. Nat. Biotechnol. 25, 1035–1044. (20) Duncan, J. S., Gyenis, L., Lenehan, J., Bretner, M., Graves, L. M., Haystead, T. A., and Litchfield, D. W. (2008) An unbiased evaluation of CK2 inhibitors by chemoproteomics: characterization of inhibitor effects on CK2 and identification of novel inhibitor targets. Mol. Cell. Proteomics 7, 1077–1088. (21) Brehmer, D., Godl, K., Zech, B., Wissing, J., and Daub, H. (2004) Proteome-wide identification of cellular targets affected by bisindolylmaleimide-type protein kinase C inhibitors. Mol. Cell. Proteomics 3, 490–500. (22) Graves, P. R., Kwiek, J. J., Fadden, P., Ray, R., Hardeman, K., Coley, A. M., Foley, M., and Haystead, T. a J. (2002) Discovery of novel targets of quinoline drugs in the human purine binding proteome. Mol. Pharmacol. 62, 1364–1372. (23) Kwiek, J. J., Haystead, T. A. J., and Rudolph, J. (2004) Kinetic mechanism of quinone oxidoreductase 2 and its inhibition by the antimalarial quinolines. Biochemistry 43, 4538–4547. (24) Wang, Z., Hsieh, T. C., Zhang, Z., Ma, Y., and Wu, J. M. (2004) Identification and purification of resveratrol targeting proteins using immobilized resveratrol affinity chromatography. Biochem. Biophys. Res. Commun. 323, 743–749. (25) Buryanovskyy, L., Fu, Y., Boyd, M., Ma, Y., Hsieh, T. C., Wu, J. M., and Zhang, Z. (2004) Crystal structure of quinone reductase 2 in complex with resveratrol. Biochemistry 43, 11417–11426. (26) Nosjean, O., Ferro, M., Coge, F., Beauverger, P., Henlin, J. M., Lefoulon, F., Fauchere, J. L., Delagrange, P., Canet, E., and Boutin, J. a. (2000) Identification of the melatonin-binding site MT3 as the quinone reductase 2. J. Biol. Chem. 275, 31311–31317. (27) Leung, K. K. K., and Shilton, B. H. (2013) Chloroquine binding reveals flavin redox switch function of quinone reductase 2. J. Biol. Chem. 288, 11242–11251. 29 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 38

(28) Senda, T., Senda, M., Kimura, S., and Ishida, T. (2009) Redox control of protein conformation in flavoproteins. Antioxid. Redox Signal. 11, 1741–1766. (29) Foster, C. E., Bianchet, M. a, Talalay, P., Zhao, Q., and Amzel, L. M. (1999) Crystal structure of human quinone reductase type 2, a metalloflavoprotein. Biochemistry 38, 9881–9886. (30) Hegde, A. H., and Seetharamappa, J. (2014) Fluorescence and circular dichroism studies on binding and conformational aspects of an anti-leukemic drug with DNA. Mol. Biol. Rep. 41, 67–71. (31) Hossain, M., Giri, P., and Kumar, G. S. (2008) DNA intercalation by quinacrine and methylene blue: a comparative binding and thermodynamic characterization study. DNA Cell Biol. 27, 81–90. (32) Sakore, T. D., Jain, S. C., Tsai, C. C., and Sobell, H. M. (1977) Mutagen-nucleic acid intercalative binding: structure of a 9-aminoacridine: 5-iodocytidylyl(3’-5')guanosine crystalline complex. Proc. Natl. Acad. Sci. U. S. A. 74, 188–192. (33) Nolan, K. A., Caraher, M. C., Humphries, M. P., Bettley, H. A.-A., Bryce, R. a, and Stratford, I. J. (2010) In silico identification and biochemical evaluation of novel inhibitors of NRH:quinone oxidoreductase 2 (NQO2). Bioorg. Med. Chem. Lett. 20, 7331–6. (34) Nolan, K. A., Humphries, M. P., Bryce, R. A., and Stratford, I. J. (2010) Imidazoacridin-6-ones as novel inhibitors of the quinone oxidoreductase NQO2. Bioorg. Med. Chem. Lett. 20, 2832–2836. (35) Lerman, L. S. (1961) Structural considerations in the interaction of DNA and acridines. J. Mol. Biol. 3, 18–30. (36) Ulitzur, S., and Weiser, I. (1981) Acridine dyes and other DNA-intercalating agents induce the luminescence system of luminous bacteria and their dark variants. Proc. Natl. Acad. Sci. U. S. A. 78, 3338–3342. (37) Böhner, R., and Hagen, U. (1977) Action of intercalating agents on the activity of DNA polymerase I. Biochim. Biophys. Acta 479, 300–310. (38) McCann, J., and Ames, B. N. (1976) Detection of carcinogens as mutagens in the Salmonella/microsome test: assay of 300 chemicals: discussion. Proc. Natl. Acad. Sci. U. S. A. 73, 950–954. (39) Ferguson, L. R., and Denny, W. a. (2007) Genotoxicity of non-covalent interactions: DNA intercalators. Mutat. Res. 623, 14–23. (40) Cascorbi, I. (2011) P-glycoprotein: Tissue distribution, substrates, and functional consequences of genetic variations, in Handbook of Experimental Pharmacology, pages 261–283. (41) Leung, K. K. K., Litchfield, D. W., and Shilton, B. H. (2012) Flavin adenine dinucleotide content of quinone reductase 2: analysis and optimization for structure-function studies. Anal. Biochem. 420, 84–89. (42) Knox, R. J., Jenkins, T. C., Hobbs, S. M., Chen, S., Melton, R. G., and Burke, P. J. (2000) Bioactivation of 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) by human NAD(P)H quinone oxidoreductase 2: a novel co-substrate-mediated antitumor prodrug therapy. Cancer Res. 60, 4179–4186. (43) Leung, K. K. K., and Shilton, B. H. (2015) Quinone reductase 2 is an adventitious target of 30 ACS Paragon Plus Environment

Page 31 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

protein kinase CK2 inhibitors TBBz (TBI) and DMAT. Biochemistry 54, 47–59. (44) Leslie, A. G. W., and Powell, H. R. (2007) Evolving Methods for Macromolecular Crystallography. Evol. Methods Macromol. Crystallogr. (Read, R. J., and Sussman, J. L., Eds.) 245, 41–51. (45) Evans, P. (2006) Scaling and assessment of data quality. Acta Crystallogr. D. Biol. Crystallogr. 62, 72–82. (46) Adams, P. D., Afonine, P. V, Bunkóczi, G., Chen, V. B., Davis, I. W., Echols, N., Headd, J. J., Hung, L.-W., Kapral, G. J., Grosse-Kunstleve, R. W., McCoy, A. J., Moriarty, N. W., Oeffner, R., Read, R. J., Richardson, D. C., Richardson, J. S., Terwilliger, T. C., and Zwart, P. H. (2010) PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D. Biol. Crystallogr. 66, 213–221. (47) Durrant, J. D., De Oliveira, C. A. F., and McCammon, J. A. (2011) POVME: An algorithm for measuring binding-pocket volumes. J. Mol. Graph. Model. 29, 773–776. (48) Durrant, J. D., Votapka, L., Sørensen, J., and Amaro, R. E. (2014) POVME 2.0: An Enhanced Tool for Determining Pocket Shape and Volume Characteristics. J. Chem. Theory Comput. 10, 5047–5056. (49) Jain, S. C., and Sobell, H. M. (1984) Visualization of drug-nucleic acid interactions at atomic resolution. VII. Structure of an ethidium/dinucleoside monophosphate crystalline complex, ethidium: uridylyl(3’-5') adenosine. J. Biomol. Struct. Dyn. 1, 1161–1177. (50) Skehan, P., Storeng, R., Scudiero, D., Monks, A., McMahon, J., Vistica, D., Warren, J. T., Bokesch, H., Kenney, S., and Boyd, M. R. (1990) New colorimetric cytotoxicity assay for anticancer-drug screening. J. Natl. Cancer Inst. 82, 1107–1112. (51) LIAO, S., DULANEY, J. T., and WILLIAMS-ASHMAN, H. G. (1962) Purification and properties of a flavoprotein catalyzing the oxidation of reduced ribosyl nicotinamide. J. Biol. Chem. 237, 2981– 2987. (52) Nolan, K. A., Humphries, M. P., Barnes, J., Doncaster, J. R., Caraher, M. C., Tirelli, N., Bryce, R. a., Whitehead, R. C., and Stratford, I. J. (2010) Triazoloacridin-6-ones as novel inhibitors of the quinone oxidoreductases NQO1 and NQO2. Bioorganic Med. Chem. 18, 696–706. (53) Velazquez-Campoy, A., and Freire, E. (2006) Isothermal titration calorimetry to determine association constants for high-affinity ligands. Nat. Protoc. 1, 186–191. (54) Reddy, B. S., Seshadri, T. P., Sakore, T. D., and Sobell, H. M. (1979) Visualization of drug-nucleic acid interactions at atomic resolution. V. Structure of two aminoacridine--dinucleoside monophosphate crystalline complexes, proflavine--5-iodocytidylyl (3’-5') guanosine and acridine orange--5-iodocytidylyl (3'-5') guanosin. J. Mol. Biol. 135, 787–812. (55) Akhter, M. Z., and Rajeswari, M. R. (2014) Interaction of doxorubicin with a regulatory element of hmga1 and its in vitro anti-cancer activity associated with decreased HMGA1 expression. J. Photochem. Photobiol. B Biol. 141, 36–46. (56) Nafisi, S., Saboury, A. A., Keramat, N., Neault, J.-F., and Tajmir-Riahi, H.-A. (2007) Stability and structural features of DNA intercalation with ethidium bromide, acridine orange and methylene 31 ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 38

blue. J. Mol. Struct. 827, 35–43. (57) Li, N., Ma, Y., Yang, C., Guo, L., and Yang, X. (2005) Interaction of anticancer drug mitoxantrone with DNA analyzed by electrochemical and spectroscopic methods. Biophys. Chem. 116, 199–205. (58) Bradshaw-Pierce, E. L., Pitts, T. M., Tan, A.-C., McPhillips, K., West, M., Gustafson, D. L., Halsey, C., Nguyen, L., Lee, N. V., Kan, J. L. C., Murray, B. W., and Eckhardt, S. G. (2013) Tumor PGlycoprotein Correlates with Efficacy of PF-3758309 in in vitro and in vivo Models of Colorectal Cancer. Front. Pharmacol. 4, 1–11. (59) Bachur, N. R., and Gee, M. (1971) Daunorubicin metabolism by rat tissue preparations. J. Pharmacol. Exp. Ther. 177, 567–572. (60) Asbell, M. A., Schwartzbach, E., Bullock, F. J., and Yesair, D. W. (1972) Daunomycin and adriamycin metabolism via reductive glycosidic cleavage. J. Pharmacol. Exp. Ther. 182, 63–9. (61) Liao, S., Dulaney, J. T., and Williams-Ashman, H. G. (1962) Purification and properties of a flavoprotein catalyzing the oxidation of reduced ribosyl nicotinamide. J. Biol. Chem. 237, 2981– 2987. (62) Chiarugi, A., Dölle, C., Felici, R., and Ziegler, M. (2012) The NAD metabolome--a key determinant of cancer cell biology. Nat. Rev. Cancer 12, 741–752. (63) Benoit, C.-E., Bastianetto, S., Brouillette, J., Tse, Y., Boutin, J. a, Delagrange, P., Wong, T., Sarret, P., and Quirion, R. (2010) Loss of quinone reductase 2 function selectively facilitates learning behaviors. J. Neurosci. 30, 12690–12700. (64) Iskander, K., Paquet, M., Brayton, C., and Jaiswal, A. K. (2004) Deficiency of NRH:quinone oxidoreductase 2 increases susceptibility to 7,12-dimethylbenz(a)anthracene and benzo(a)pyrene-induced skin carcinogenesis. Cancer Res. 64, 5925–5928. (65) Long, D. J., Iskander, K., Gaikwad, A., Arin, M., Roop, D. R., Knox, R., Barrios, R., and Jaiswal, A. K. (2002) Disruption of dihydronicotinamide riboside:quinone oxidoreductase 2 (NQO2) leads to myeloid hyperplasia of bone marrow and decreased sensitivity to menadione toxicity. J. Biol. Chem. 277, 46131–46139. (66) Reddy, P. V. N., Jensen, K. C., Mesecar, A. D., Fanwick, P. E., and Cushman, M. (2012) Design, synthesis, and biological evaluation of potent quinoline and pyrroloquinoline ammosamide analogues as inhibitors of quinone reductase 2. J. Med. Chem. 55, 367–377. (67) Dunstan, M. S., Barnes, J., Humphries, M., Whitehead, R. C., Bryce, R. a, Leys, D., Stratford, I. J., and Nolan, K. A. (2011) Novel inhibitors of NRH:quinone oxidoreductase 2 (NQO2): crystal structures, biochemical activity, and intracellular effects of imidazoacridin-6-ones. J. Med. Chem. 54, 6597–6611.

32 ACS Paragon Plus Environment

Page 33 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Leung and Shilton Figure 1

Ethidium (Homidium)

Acridine orange

Doxorubicin (Adriamycin)

Methylene blue

Mitoxantrone

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 38

Leung and Shilton Figure 2

A

B

ACS Paragon Plus Environment

Page 35 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Leung and Shilton Figure 3

A

G174

B

W105

G174

D117

W105

D117

F178

F178

T71 N161

T71

N161

G68

C

G68

D

G174

G174

W105

W105

F178

F178 N161 N161

M154 M154 E193

E

G174 W105

F178

N161

M154

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 38

Leung and Shilton Figure 4

A 13.5 Å

polar hydrophobic

3.7 Å polar

nucleotide

2.5 Å

7.9 Å

B polar hydrophobic

polar

13.9 Å

2.7 Å

FAD FAD

FAD

C

D E193

ACS Paragon Plus Environment

Page 37 of 38

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Leung and Shilton Figure 5

A

B

C

ACS Paragon Plus Environment

DNA

DNA Biochemistry Ethidium

1 2 3 4

Page 38 of 38

Acridine orange

ACS Paragon Plus Environment

ToC graphics

NQO2

NQO2